Beginning this summer the 500 largest US public mpanies start reporting their business performance data using technology that enables a whole new level of transparency. That technology is a data reporting language called XBRL – eXtensible Business Reporting Language. Perhaps XBRL uld also be used to provide transparency in how bank bailout and enomic stimulus money is spent.

For the last five years Sarbanes-Oxley requirements in the United States have resulted in voluminous additional reporting on performance of public mpanies, so how me we didn’t use this data to better analyze what they were doing and avoid or at least anticipate the stock market meltdown? “Requiring people to report more and more doesn’t solve the problem; it’s how you report the data that solves the problem.” That’s the answer Philip Moyer gave to my question. Philip is the CEO of EDGAR Online (EDGAR stands for electronic data gathering and reporting). His mpany is the number one source of financial reporting data in the US outside of the Securities and Exchange mission (SEC).

Regulations are Good, but Transparency is Better

To illustrate his answer Philip noted that largely in response to new regulations the annual 10-K report filed by Citigroup grew from 265 pages in 1999 to 1,376 pages in 2008, yet this five-fold increase in volume has actually made it harder to see what is really happening. “The more reporting requirements we mandate, the more text and tables are used and they are different from one mpany to another,” he said. He explained that in 1933 (after the last big stock market meltdown) the SEC was created and they required all public mpanies to report four basic numbers: 1) assets; 2) liabilities; 3) revenues; and 4) profits. Those numbers gave investors a way to judge for themselves how well mpanies were doing and what their stock was worth. But over the years regulations have gotten more mplex as businesses became more mplex. Now in the worst financial crisis since 1933, people cannot easily figure out how well mpanies are doing because their important numbers are buried in thousands and thousands of pages of text and presented in tables that are not standard from one mpany to another.

In addition to figuring out how an individual mpany such as a bank is doing, it’s just as important to understand how a mpany is doing in relation to other mpanies in its industry because that provides a meaningful ntext in which to judge performance. Everything nforms to a bell curve and establishing the relevant performance bell curve for an industry is a big part of providing transparency. The performance bell curve for an industry shows which mpanies are leading, which mpanies are performing close to the industry norm and which mpanies are lagging. When you see which mpanies are leading and lagging then you can make sense of individual mpany performance.

Philip cited a CFA Institute (Chartered Financial Analysts) study done in 2007 that revealed more than 70% of financial analysts hand-enter data from public mpany 10-K and 10-Q reports into spreadsheets in order to do analysis of those mpanies. That means they first have to plow through thousands of pages of text to pull out relevant numbers and then they have to do their best to standardize the numbers from different mpanies so that they represent the same measurements and can be mpared from one mpany to another. Between the work to find the relevant data, standardize it, and manually enter it into spreadsheets and databases, there are lots of opportunities for error and all of thiswork is so time nsuming that most financial analysts spend most of their time doing low value data entry work instead of the high value analysis they are trained to do. How can we use technology to help?

There is an Answer to This Problem

Using XBRL mpanies can tag and report their performance data using a standard taxonomy of data definitions. All the data elements in a 10-K or 10-Q report are vered by standard XBRL data tags. XBRL is an open data specification and there are currently some 15,000 data tags in its taxonomy. XBRL was originally proposed in 1997 and in 1999 it was approved by the international body that sets standards for XBRL - XBRL International. EDGAR Online was a founding member of this international erning body.

XBRL is already used by numerous financial and ernmental anizations around the world. Stock exchanges in China, Japan, Singapore and South Korea all mandate XBRL data. Europe has developed an array of ernment-wide and cross-border applications that share nsistently structured XBRL data. In the US the 500 largest public mpanies are using it (they represent 86% of total US market capitalization) and next about 1,500 additional mpanies (representing another 12% of total market cap) will start using XBRL. The last 10,000 public mpanies (mprising the final 2% of market cap) will use XBRL for their reporting by the end of 2011. If mpanies are not able to deliver their data in XBRL on their own then EDGAR Online and numerous other nsulting firms and technology vendors can perform this data translation for them.

Once this happens, the data from all public mpanies in the US can flow automatically into the spreadsheets and performance models of financial analysts and all those analysts can spend their time doing analysis instead of data entry. Industry performance bell curves will be much easier to see and performance of individual mpanies will also be much easier to judge. This transparency will go a long way toward restoring trust in our financial markets.

QUESTION: Since there is already an existing data reporting language that nforms to open international standards being used by mpanies and ernments around the world why would we not also use this same technology to report on how bank bailout money and US federal ernment enomic stimulus money is being spent?

At present banks in the US are not required to report on how they use their bailout funds (see last letter to Treasury Secretary Geithner from ngressional Oversight Panel at end of this report) and it is unclear what is actually happening with this money. As the stimulus money starts being spent over the next 12 months there is an equally strong need to know how that money is being spent. Without real transparency, the trillions of dollars of bailout and stimulus money will not be spent as wisely and effectively as we taxpayers would like.

QUESTION: Is there any valid reason not to use XBRL to provide transparency to monitor the largest US ernment spending programs since World War II?